Apple has detailed a set of new security and privacy controls coming to macOS 27.3, following an announcement made in October, as the company moves to address risks posed by AI agents and other applications. The update introduces expanded permission controls that will require users to explicitly authorize access to sensitive data and system resources, alongside new warning prompts designed to flag potentially risky behavior by automated tools. Apple said the features are intended to give users greater visibility into how applications, particularly AI-driven agents capable of acting on a user's behalf, interact with their devices and personal information.
The company also outlined transparency measures that will notify users when AI agents or other apps attempt actions that could affect privacy, such as accessing files, controlling other applications, or transmitting data. Apple framed the changes as a proactive response to the growing complexity of software that operates with increasing autonomy, noting that traditional permission models were not designed for tools that can chain multiple actions together. The controls will roll out with macOS 27.3, though Apple did not specify a release date. Privacy advocates have broadly welcomed the added safeguards, while developers are assessing how the stricter requirements may affect app functionality and user experience.
Independent reporting synthesized from primary wires and global correspondents.